Robot Ideas

[cavetroll]

Me and a few friends (paulm, jdeman, among others) are planning on entering Robofest (http://www.robofest.net/). We would be entering the exhibition category.

Our problem is in coming up with ideas. When we attempt to brainstorm, we always come up with ideas that would be far too hard, or far too easy for our abilities (The other result is that we spend half an hour coming up with crazy uranium powered terminator robots). If anyone has any suggestions as to what we should build, we would love to hear about it.

[Insectoid]

Build a checker- or chess-playing robot. That would have to have some interesting AI, and the mechanical bit would be interesting.

[paulm]

insectoid wrote:

Build a checker- or chess-playing robot. That would have to have some interesting AI, and the mechanical bit would be interesting.

We thought of that, but its been done before unfortunately...

Connect 4 perhaps?

[Tony]

Does it have to be original? It's been done a bunch of times, but Rubik's Cube solving robot is always interesting to see. If you want to level up the difficulty, it could take in a specialty 4x4x4 cube or larger.

What kind of a difficulty range are you guys in? What's too easy/hard? What kind of materials are you working with?

[Vermette]

Second on the question of what hardware is available to you... I'm guessing an NXT?

[Insectoid]

I've always wanted one of those. I have the original mindstorms kit, which is incompatible with XP, and the USB ports on my 98 don't work any more. Hmm, maybe it'll run in a win2k bottle on crossover...

[paulm]

Vermette @ Sat Jan 10, 2009 3:49 am wrote:

Second on the question of what hardware is available to you... I'm guessing an NXT?

Nope; were planning on working with a laptop and an arduino to drive servos (although the arduino could certainly handle some simpler ideas i sure). I plan on buying some servos, sensors, etc according to what idea we choose.

I like the rubik's cube idea, trying to think of an orientation for the servo's though...

EDIT:

the particular model of the arduino we are planning on working with is the ArduinoDuemilanove It's based around the ATmega168 chip which has 1 KB of SRAM, 512 bytes of EEPROM and 14 k of usable ROM @ 16 hertz, if not used for the actual AI in a robot, it can at least be used for driving servos/motors (plus i own an acer aspire one if we need a laptop to drive the arduino)

[cavetroll]

Just wanted to put in my two cents. I think Paulm did a good job answering the hardware questions.

As for the originality of our design, it doesn't have to be something that has never been done before, but we would prefer to work on something that isn't overdone.

Concerning our skills, although no one on the team has actually built a robot before, we all have a fair bit of experience in our respective categories. I know that paulm has done a fair bit of electrical and mechanical work, just not with the software to interface with it. On the other hand, I have a reasonable amount of experience programming, although never with interfacing with hardware. I think the other group members are about the same.

So to sum it up, any project we attempt will bring new challenges. The idea would be to have an project that will challenge us to improve our skills, while not having to work 24/7 from now till April on it.

Also, Thanks Tony and Insectiod for the ideas.

[md]

can it be a vehicle?

If you can make a robot which will map out rooms, maybe using sonar or lidar; that would be pretty awesome.

[paulm]

md @ Sun Jan 11, 2009 1:55 am wrote:

can it be a vehicle?

If you can make a robot which will map out rooms, maybe using sonar or lidar; that would be pretty awesome.

of course it can be mobile

awesome idea, added to our list for sure as a practial robot.